{"thread":{"id":"18498","subject":"GIT BUG? GIT occasionally redownloads its entire data set","startedAt":"2009-03-24T00:45:43Z","lastAt":"2009-03-24T09:15:29Z","messageCount":5,"participants":["David Howells","Junio C Hamano","Dmitry Potapov"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"109126","messageId":"28707.1237855543@redhat.com","threadId":"18498","inReplyTo":null,"subject":"GIT BUG? GIT occasionally redownloads its entire data set","fromName":"David Howells","fromEmail":"dhowells@redhat.com","sentAt":"2009-03-24T00:45:43Z","receivedAt":"2009-03-24T00:45:43Z","isPatch":false,"sender":{"key":"dhowells@redhat.com","avatar":null},"body":"\nHi,\n\nI'm seeing a possible bug in which GIT, in which I issue a git-pull command on\na tree that was okay the day before, GIT reports that there are \"no common\ncommits\" and then redownloads the entire tree (as far as I can tell).\n\nThis happened just now when I did I git-pull on a GIT repository that I think\nwent from Monday morning's head:\n\n\thttp://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=59fcbddaff6f862cc1584b488866d9c4a5579085\n\nto v2.6.29. I've attached the command output for reference.\n\nI've seen this a number of times. It has been suggested this usually happens\nwhen a new version tag is committed, but I wouldn't normally see that as the\ngit-pull to my base repository is normally run by crond.\n\nThis bug is annoying as it can eat a big chunk out of the download quota set by\nmy ISP.\n\nDavid\n---\nwarthog>git-pull\nremote: Counting objects: 447, done.\nremote: Compressing objects: 100% (57/57), done.\nremote: Total 266 (delta 216), reused 254 (delta 207)\nReceiving objects: 100% (266/266), 37.94 KiB, done.\nResolving deltas: 100% (216/216), completed with 105 local objects.\nwarning: no common commits\nremote: Counting objects: 1075436, done.\nremote: Compressing objects: 100% (176743/176743), done.\nremote: Total 1075436 (delta 896164), reused 1072243 (delta 893708)\nReceiving objects: 100% (1075436/1075436), 260.71 MiB | 307 KiB/s, done.\nResolving deltas: 100% (896164/896164), done.\nFrom git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6\n * [new tag] v2.6.29 -> v2.6.29\nUpdating 59fcbdd..8e0ee43\nFast forward\n Documentation/filesystems/ext2.txt | 5 +-\n Documentation/filesystems/proc.txt | 7 +++\n Makefile | 7 ++-\n arch/powerpc/kernel/head_32.S | 9 ++++\n arch/sparc/include/asm/pil.h | 1 +\n arch/sparc/kernel/kgdb_64.c | 2 +-\n arch/sparc/kernel/pci_common.c | 2 +-\n arch/sparc/kernel/ttable.S | 7 +++-\n arch/sparc/mm/ultra.S | 24 +----------\n drivers/dca/dca-sysfs.c | 21 ++++++++++\n drivers/isdn/gigaset/bas-gigaset.c | 16 ++++++-\n drivers/net/Kconfig | 2 +-\n drivers/net/benet/be.h | 1 +\n drivers/net/benet/be_main.c | 60 ++++++++++++++++------------\n drivers/net/bnx2.c | 12 +++---\n drivers/net/bonding/bond_main.c | 25 ++++++++++--\n drivers/net/dm9000.c | 6 ++-\n drivers/net/dnet.c | 6 +-\n drivers/net/sh_eth.c | 20 ++++++++-\n drivers/net/sh_eth.h | 4 +-\n drivers/net/smsc911x.c | 4 ++\n drivers/net/sungem.c | 9 ++--\n drivers/net/tulip/tulip_core.c | 45 ++++++++++++++--------\n drivers/net/ucc_geth.c | 34 +++++-----------\n drivers/net/ucc_geth.h | 3 +-\n drivers/net/virtio_net.c | 3 +\n drivers/net/wireless/ath9k/ath9k.h | 4 +-\n drivers/net/wireless/ath9k/core.h | 33 ++++++++++++++++\n drivers/net/wireless/ath9k/hw.c | 22 ++++++++++-\n drivers/net/wireless/ath9k/main.c | 1 +\n drivers/net/wireless/zd1211rw/zd_mac.c | 8 +++-\n drivers/sbus/char/openprom.c | 1 +\n include/linux/dca.h | 20 ++++++++++\n include/net/netfilter/nf_conntrack_core.h | 3 +-\n kernel/signal.c | 8 ++++\n net/core/dev.c | 7 ++-\n net/ipv4/ip_fragment.c | 3 +-\n net/ipv6/netfilter/nf_conntrack_reasm.c | 8 ++--\n net/ipv6/reassembly.c | 7 +--\n net/ipv6/sit.c | 2 +-\n net/mac80211/tx.c | 2 +\n net/netfilter/nf_conntrack_core.c | 2 +-\n net/netfilter/nf_conntrack_netlink.c | 1 +\n net/netfilter/nf_conntrack_proto_tcp.c | 4 +-\n net/sctp/endpointola.c | 3 +-\n net/wireless/Kconfig | 10 +++++\n net/wireless/lib80211_crypt_ccmp.c | 2 +\n net/wireless/lib80211_crypt_tkip.c | 4 ++\n net/wireless/nl80211.c | 11 +++++\n 49 files changed, 351 insertions(+), 150 deletions(-)\nwarthog>\n"},{"id":"109149","messageId":"7vskl34qc9.fsf@gitster.siamese.dyndns.org","threadId":"18498","inReplyTo":"28707.1237855543@redhat.com","subject":"Re: GIT BUG? GIT occasionally redownloads its entire data set","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2009-03-24T02:17:10Z","receivedAt":"2009-03-24T02:17:10Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"David Howells writes:\n\n> to v2.6.29. I've attached the command output for reference.\n\nVersion of git used, and the .git/config ([remote \"origin\"] section)?\n\n> warthog>git-pull\n> remote: Counting objects: 447, done.\n> remote: Compressing objects: 100% (57/57), done.\n> remote: Total 266 (delta 216), reused 254 (delta 207)\n> Receiving objects: 100% (266/266), 37.94 KiB, done.\n> Resolving deltas: 100% (216/216), completed with 105 local objects.\n> warning: no common commits\n> remote: Counting objects: 1075436, done.\n> remote: Compressing objects: 100% (176743/176743), done.\n> remote: Total 1075436 (delta 896164), reused 1072243 (delta 893708)\n> Receiving objects: 100% (1075436/1075436), 260.71 MiB | 307 KiB/s, done.\n> Resolving deltas: 100% (896164/896164), done.\n>>From git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6\n\nThis is intereseting because it shows that two separate transfer happened\nback to back, which I do not think is what recent git does.\n\nWe used to first grab the objects reachable from primary refs (presumably\n[remote \"origin\"] fetch = refs/heads/*:refs/remotes/origin/*) and then\nturned around to ask for tags that possibly point at commits we freshly\nobtained during the first transfer (i.e. the ones reachable from the tip\nof \"master\"). In such an ancient version I vaguely recall some bugs fixed\nbefore we stopped doing this two-stage transfer altogether, but I do not\nrecall offhand if a bug that loses track of what it just downloaded (which\nis what your description suggests) was among them.\n\nRings a bell, anybody?\n"},{"id":"109150","messageId":"37fcd2780903231953pdfaa679r8a680f64ee692c8d@mail.gmail.com","threadId":"18498","inReplyTo":"7vskl34qc9.fsf@gitster.siamese.dyndns.org","subject":"Re: GIT BUG? GIT occasionally redownloads its entire data set","fromName":"Dmitry Potapov","fromEmail":"dpotapov@gmail.com","sentAt":"2009-03-24T02:53:26Z","receivedAt":"2009-03-24T02:53:26Z","isPatch":false,"sender":{"key":"dpotapov@gmail.com","avatar":"https://avatars.githubusercontent.com/u/6568595?v=4"},"body":"On Tue, Mar 24, 2009 at 5:17 AM, Junio C Hamano wrote:\n>\n> Rings a bell, anybody?\n\nMaybe this:\nhttp://article.gmane.org/gmane.comp.version-control.git/75201\n\nIt was fixed in Git 1.5.5 if I am not mistaken.\n\n\nDmitry\n"},{"id":"109156","messageId":"7vmybb4fwh.fsf@gitster.siamese.dyndns.org","threadId":"18498","inReplyTo":"37fcd2780903231953pdfaa679r8a680f64ee692c8d@mail.gmail.com","subject":"Re: GIT BUG? GIT occasionally redownloads its entire data set","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2009-03-24T06:02:38Z","receivedAt":"2009-03-24T06:02:38Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Dmitry Potapov writes:\n\n> On Tue, Mar 24, 2009 at 5:17 AM, Junio C Hamano wrote:\n>>\n>> Rings a bell, anybody?\n>\n> Maybe this:\n> http://article.gmane.org/gmane.comp.version-control.git/75201\n>\n> It was fixed in Git 1.5.5 if I am not mistaken.\n\nThanks. Distributed memory works very well ;-)\n"},{"id":"109191","messageId":"32484.1237886129@redhat.com","threadId":"18498","inReplyTo":"37fcd2780903231953pdfaa679r8a680f64ee692c8d@mail.gmail.com","subject":"Re: GIT BUG? GIT occasionally redownloads its entire data set","fromName":"David Howells","fromEmail":"dhowells@redhat.com","sentAt":"2009-03-24T09:15:29Z","receivedAt":"2009-03-24T09:15:29Z","isPatch":false,"sender":{"key":"dhowells@redhat.com","avatar":null},"body":"Dmitry Potapov wrote:\n\n> It was fixed in Git 1.5.5 if I am not mistaken.\n\nwarthog>rpm -q git\ngit-1.5.4.3-3.fc8\n\nIt looks like I'm using a version of GIT that's just a little bit too old.\n\nDavid\n"}]}